71446 research outputs found
Sort by
Transfer by Design: Learning in the Flow of Work During Complex and Changing Times
No full textRapid technological advancements and global disruptions, as witnessed in the COVID-19 pandemic, are driving an increasingly volatile, uncertain, complex, and ambiguous world of work. In-demand jobs change rapidly, as do the tasks within occupations and the tools, mediums, and environments in which work takes place. While these shifts present significant challenges, they also offer immense opportunities, necessitating that both new and existing workers develop adaptability and resilience. However, many traditional approaches to workforce learning require substantial time and resources, often emphasizing narrow, role-specific skills. These approaches vary widely in quality and can lead to unintended consequences, highlighting the need for more nuanced learning strategies.
This dissertation comprises four independent yet interrelated studies conducted within two multinational consulting firms. Drawing on semi-structured interviews, Studies One and Two examine how individuals navigate modern apprenticeship programs. Study One identifies key pedagogical features of these programs in hybrid contexts, emphasizing learning through developmental relationships. Study Two highlights the networked nature of learning within modern apprenticeships and demonstrates that proximity to experienced mentors enhances apprenticeship experiences. Both studies underscore the importance of strong developmental relationships characterized by frequent communication, worked examples, and shared thinking.
Studies Three and Four explore the experiences of mid-career consultants who frequently undergo rapid role transitions. Study Three, based on extensive interviews across two consulting firms, reveals that learning transfer, the ability to apply learning in one context to a novel one, is not an isolated process but a dynamic one embedded within interpersonal relationships, developmental contexts, and complex cultural environments. Study Four employs design-based research methods to identify features and behaviors that enable workers to transfer their existing knowledge, skills, and abilities to new
contexts. It finds that the presence of pro-transfer features and behaviors in relationships with experienced mentors is associated with improved performance on transfer tasks and greater confidence in one’s ability to transfer learning.
Together, these studies suggest that fostering adaptability and resilience in the workforce requires the intentional design, enactment, and sustainment of positive developmental relationships with experienced mentors in the workplace.Educatio
Patterns of Place: Housing Supply, Racial Segregation, and the Suburbanization of Immigration
No full textPlace is central to understanding inequality. However, recent decades have seen significant shifts in the urban layout of the United States. In this dissertation, I address changes in housing and immigration through a mixed-methods approach combining causal inference, decomposition methods, and qualitative approaches. In particular, I analyze three recent spatial dynamics: declines in housing development, variation in racial integration, and the suburbanization of immigration.
Chapter Two examines Massachusetts’ Chapter 40B, a state law designed to override exclusionary zoning and expedite affordable housing projects, to analyze how informal opposition shapes housing outcomes. Housing affordability debates often focus on zoning regulations, but informal opposition through bargaining, delay, and deterrence also plays a critical role in restricting development. Using data collected from observations and public meeting records of 48 40B projects, I identify three key dynamics: density primacy, where opponents link concerns to project size; issue shifting, where objections evolve to sustain opposition; and policy learning, where opponents refine arguments to align with regulatory constraints. These strategies allow local actors to extract concessions and reshape developments even when formal mechanisms for denial are unavailable. Findings highlight the limitations of zoning reform alone in addressing housing shortages and underscore the need for policy solutions that mitigate procedural barriers to housing production. While public meetings are intended to facilitate community input, they often amplify opposition in ways that hinder new housing construction.
In Chapter Three, I investigate whether increasing housing development reduces Black-White residential segregation. Debates over housing supply, from NIMBY opposition to YIMBY advocacy, have focused largely on the economic consequences of permitting new housing, but their broader social effects remain underexplored. Using a panel of metropolitan areas from 1990 to 2020, I find that metros permitting more housing experienced larger declines in segregation. To strengthen causal claims, I employ an instrumental variable approach leveraging geologic constraints on buildability. I also show that within metro areas, towns permitting more housing saw greater reductions in segregation, reinforcing the link between housing supply and integration. These findings highlight the broader social consequences of housing development policies, suggesting that increasing housing supply may be a key lever for fostering racial equity.
In Chapter Four, I identify and question three assumptions about the suburbanization of immigration. The majority of immigrants in the U.S. now live in the suburbs of major metropolitan areas. Recent work on this phenomenon has focused on the outcomes of immigrant suburbanization, examining how immigrants encounter and make sense of different residential contexts. However, I argue that the literature has largely neglected to investigate the underlying process behind immigrant suburbanization. I analyze three assumptions related to this process: that moves to new destinations are driving suburbanization, that immigrants are suburbanizing more quickly than the U.S.-born, and that a suburb-specific mechanism is driving increases in immigrant-native neighborhood inequality. I use a series of decompositions to model each of these questions, finding evidence that largely cuts against these underlying assumptions in the field. In doing so, I show the value of situating spatial trends in immigration in the context of broader demographic trends and help clarify possible mechanisms driving immigrant suburbanization.Sociolog
Investigating PD-1 regulation of CD8+ T cell fate following acute influenza infection
No full textThe cornerstone of the adaptive immune system is the capacity to remember and rapidly respond to previously encountered pathogens. Tissue-resident memory (TRM) CD8+ T cells represent a specialized subset of memory T cells that maintain permanent residence at common barrier tissues and enable rapid response to tissue perturbation upon re-encounter with antigen. Checkpoint receptors such as PD-1 have been heavily studied in the context of chronic antigen settings but are also known to be upregulated constitutively on TRM. However, the impact of PD-1 on the differentiation and function of CD8+ TRM cells following acute contexts is not well understood.
This work explores how PD-1 signaling regulates the differentiation of CD8+ TRM cells following acute influenza infection. Genetic deletion of PD-1 in influenza-specific CD8+ T cells impaired acquisition of CD69 and CD103 – canonical TRM markers – at both effector and memory time points. Targeted deletion of an exhaustion-associated PD-1 enhancer region attenuated PD-1 expression in CD8+ T cells in flu-infected tissues, reducing tissue residency marker acquisition. Mouse models with germline or CD8-specific PD-1 loss had reduced tissue-residency marker expression within antigen-experienced CD8+ T cells. However, antibody blockade of the PD-1 pathway during priming did not recapitulate the TRM defects observed with genetic deletion. Our findings highlight PD-1's complex, context-dependent effects on CD8+ T cell fate decisions and establish its critical role in balancing protective immunity with tissue damage following respiratory viral infection.Graduate Educatio
On-chip generation and manipulation of quantum states of light in thin-film lithium niobate
Get PDFWithin the last decade, thin-film lithium niobate (TFLN) has emerged as a leading integrated photonics platform with immediate applications in classical communications. Concurrently, TFLN components tailored to the requirements of photonic quantum information processing have also garnered considerable interest. Chief among these is the quasi-phase matched (QPM) nonlinear
frequency mixer—implemented via periodic domain inversion in ferroelectric lithium niobate—which can be used to generate photon pairs, squeezed states of light, and to perform single photon frequency conversion. Here, we demonstrate progress towards realizing a spectrally separable photon pair source in lithium niobate via waveguide dispersion engineering—a technique uniquely enabled by sub-wavelength optical mode confinement in the thin-film platform. Subsequently, we design optimize a scalable fabrication process to produce QPM TFLN devices for applications that require strict adherence to a specified operating wavelength, such as quantum frequency conversion in a quantum communications network. Finally, we explore how high-performance electro-optic devices can be combined with these nonlinear optical devices to realize a multi-functional platform in which quantum states of light can be generated and manipulated within a single, compact photonic integrated circuit.Engineering and Applied Sciences - Applied Physic
Large Scale, User-Defined Peptide and Peptide-Human Leukocyte Antigen Library for High Throughput Detection of Immunogenic Antigens
No full textTargeted immunotherapies—ranging from personalized cancer vaccines to adoptive T‐cell therapies—have revolutionized oncology by leveraging the specificity of T cells to recognize tumor‐associated peptide-bound Human Leukocyte Antigen complexes (pHLA), leading to a durable and robust T cell response against infected or tumor cells. While the process of profiling the pHLA repertoire through mass spectrometry (MS), also known as immunopeptidomics, excels at identifying abundant peptides, it routinely misses lower‐abundance antigens, noncanonical antigens, and neoantigens—antigens arising from tumor-specific somatic mutations. Conversely, existing computational models—though high‐throughput—are constrained by limited HLA allele coverage and poor incorporation of peptide stability or immunogenicity features. These gaps limit our ability to comprehensively elucidate the tumor immunopeptidome and to discover the most clinically effective targets for next‐generation, personalized immunotherapies.
To address the limitations in immunopeptidomics and the biases inherent in contemporary in silico predictors, we have devised two synergistic high-throughput platforms based on the recombinant protein expression system in Escherichia coli (E. coli): (1) Pepyrus: a method for rapid and scalable production of user-defined pure synthetic peptides and peptide spectral libraries, and (2) a method for production of recombinant pHLA from user-defined peptide libraries for targeted screening of HLA-bound peptide.
Using Pepyrus, we produced user-defined peptide libraries totalling over 100,000 patient-specific and off-the-shelf shared cancer-specific peptides and acquired extensive reference spectra through high-resolution MS. From the peptide spectral libraries, we demonstrated its ability to improve low-abundance peptide detection when used together with data-independent mass spectrometry (DIA-MS), an acquisition method which provides a comprehensive spectral map of a given sample but requires a reference spectral library for peptide identity deconvolution. Pepyrus significantly increased detection sensitivity for primary tumor samples and patient-derived cell lines, successfully recovering numerous low-abundance neoantigen, endogenous retroviral, and unannotated open-reading-frame peptides that were undetected by conventional MS methods.
Subsequently, we combined Pepyrus with recombinant HLA expression in E. coli and generated a recombinant pHLA library encompassing 10,000 peptides across 10 prevalent class I HLA alleles, systematically measuring binding at scale. All HLA alleles tested identified novel bound peptide motifs, thus potentially expanding the space of known allele-specific peptide sequences which could enhance the accuracy of prediction models to predict genuine binders. The design of the pHLA construct also allows for high-throughput screening and heat treatments, potentially allowing for systematic interrogation of features related to peptide stability, which could enhance the accuracy of predicting immunogenic epitopes within patient samples.
Collectively, these integrated pipelines amplify the scope and intricacy of immunopeptidome profiling in a user-defined and high throughput manner, facilitating the reliable identification of both shared antigens and patient-specific neoantigens for the development of personalized vaccines and adoptive T-cell therapies. Our approach establishes a robust foundation for next-generation antigen detectors and predictors and paves the way for more efficient and broadly applicable targeted immunotherapies.Biological and Biomedical Science
Positivity in Cluster Algebras and Their Generalizations
Get PDFThe theory of bf cluster algebras gives us a combinatorial framework for understanding the previously opaque nature of certain algebras. Each cluster algebra is generated by its cluster variables, which can be obtained via the recursive process of mutation. One remarkable property of cluster algebras is Laurent positivity, which means that every cluster variable can be written in a nice form; specifically, as a Laurent polynomial with positive integer coefficients in the initial cluster variables. Laurent positivity for cluster algebras unifies positivity phenomena in a variety of contexts, including Teichmuller theory, Gromov-Witten theory, string theory, and tropical geometry.
Laurent positivity was conjectured by Fomin and Zelevinsky when they introduced cluster algebras in 2002, but the proof remained elusive for over a decade. There have since been two proofs: a combinatorial approach by Lee and Schiffler, and a geometric approach by Gross, Hacking, Keel, and Kontsevich using a novel connection to scattering diagrams. Scattering diagrams themselves are powerful tools, originating from mirror symmetry, where they track how certain geometric invariants (Gromov--Witten invariants and Donaldson--Thomas invariants) change under varying stability conditions. Every cluster algebra is associated with a cluster scattering diagram that encodes algebraic relations between cluster variables, making them a useful tool in cluster algebra theory.
The work in this dissertation unifies these methods, aiming to deepen our understanding of positivity in both cluster algebras and scattering diagrams. In Chapter 3, which is joint work with Kyungyong Lee and Lang Mou, we prove positivity for generalized cluster algebras of all ranks, confirming a 2014 conjecture of Chekhov--Shapiro. We achieve this by giving a directly computable, manifestly positive, and elementary but highly nontrivial formula describing rank 2 generalized cluster scattering diagrams. This formula enumerates a new class of Dyck path objects, called tight gradings, implying positivity of the scattering diagrams in rank 2. In Chapter 4, which is joint work with Kyungyong Lee, we construct an explicit bijection between broken lines on scattering diagrams and compatible pairs on Dyck paths, which both play crucial roles in the proofs of cluster algebra positivity. In Chapter 5, we give a new expansion formula for quantum cluster variables using colored subpaths of Dyck paths, leveraging a connection we make to the compatible pair framework.Mathematic
Statistical methods for polygenic risk prediction from biobanks and genome-wide association studies
No full textPolygenic risk scores (PRS) have emerged as a promising tool to translate genomic discoveries into clinic decision making. By aggregating the effects of risk-associated genetic variants across the genome into a single number, PRS can quantify a patient’s genetic predisposition for a wide range of health outcomes. The past decade has seen an explosion of statistical methodology to build PRS prediction models, but these existing methods face limitations in prediction accuracy, computational efficiency, and generalizability across populations. In this dissertation, we present three novel approaches to tackle these challenges.
In Chapter 1, we propose ALL-Sum, an ensemble learning-based PRS method that uses summary statistics from genome-wide association studies (GWAS). ALL-Sum leverages L0L2-penalized regression and fast optimization algorithms to enable high prediction accuracy while also dramatically reducing the computational runtime and memory usage. Then, in Chapter 2, we propose SPLENDID, which models gene-by-ancestry interactions to simultaneously capture shared and heterogeneous genetic effects without categorizing individuals into discrete ancestry groups, allowing for fairer clinical implementation in diverse patient populations. Finally, in Chapter 3, we propose STELLAR, which ensembles multiple prediction modeling approaches and functional genomic annotations to flexibly estimate rare variant effects for more comprehensive genome-wide PRS.
Altogether, the work from this dissertation introduces new statistical frameworks to efficiently compute accurate PRS from large-scale genomic data. Each method demonstrates substantial improvements over the current state-of-the-art through comprehensive simulation studies and application to real data. These tools can be used to develop new prediction models for complex traits and diseases and ultimately advance precision medicine.Biostatistic
Quantum algorithms and quantum error correction with neutral atoms
No full textQuantum computers have the potential to solve certain problems exponentially faster than classical computers. However, realizing this potential in practice is a major challenge, as it requires precise control over large-scale quantum systems operating at extremely low error rates. Quantum error correction (QEC) provides a path to achieving such error rates, but its substantial resource overhead poses a significant practical barrier. In addition, identifying which computational problems offer a provable quantum advantage—and which remain fundamentally intractable—remains an open question. This thesis presents progress towards addressing both of these challenges.
The first part of this thesis describes advances that substantially reduce the resource overhead of QEC. We begin by presenting realizations of logical circuits in dynamically reconfigurable arrays of neutral atoms. By jointly decoding the logical qubits, we reduce the cost of implementing such transversal Clifford circuits by a factor proportional to the code distance. We then develop new theories of fault tolerance to extend these savings to universal quantum computation with magic state inputs. Finally, we introduce techniques for fast correlated decoding, enabling practical implementations of these improvements in experimental hardware. Collectively, these advances accelerate progress toward large-scale computation and reduce the cost of QEC by over an order of magnitude.
The second part of the thesis explores the relative power of quantum and classical computation in combinatorial optimization, a class of problems that is ubiquitous in science and engineering and foundational to the theory of computational complexity. We experimentally implement the optimized quantum adiabatic algorithm (QAA) in neutral atom arrays and observe evidence of a superlinear speedup over classical simulated annealing on certain hard problem instances. To interpret these results, we develop a theoretical framework to compare the performance of QAA with a broad class of classical Markov chain Monte Carlo algorithms. We identify conditions under which a quantum quadratic speedup is achievable and propose modifications to the QAA to reliably realize this advantage. Together, these contributions advance both the practical implementation and theoretical understanding of quantum algorithms.Physic
Membrane-Electrolyte System Studies for Aqueous Redox Flow Reactors
Get PDFElectrochemical flow reactors are promising for electrochemistry at scale. Reactants are flowed continuously through typically porous electrodes where redox occurs, and ions transport through the electrolyte and typically at least one ion exchange membrane to provide current in the electrochemical circuit and balance charge. Ion exchange membranes are made of charged polymers that take up solvent to create ionically conductive pathways while blocking bulk mixing of adjacent electrolyte solutions. This thesis examines a crucial interface in electrochemical flow reactors, where the membrane and contacting electrolyte interact and exchange chemical species, and we find that the interaction of membrane and electrolyte governs the structure and transport of the membrane phase, which affects device scale performance. In redox flow batteries, the membrane must block crossover of reactants while enabling high conductivity. A combination of reactant size and charge effects influence permeation rates through charged membranes, with charge exerting especially strong influence under dilute conditions. The overall concentration and composition of the battery electrolyte influences the membrane hydration and hence transport, and we use conclusions from a systematic study of these effects to design crossover-free membrane-electrolyte systems with both commercial and novel membranes. In bipolar membranes, we find that the composition of impure strong electrolytes affects the local composition at the bipolar junction, which determines open circuit voltage and polarization behavior. Finally, we involve a series of ion exchange membranes including a bipolar membrane in a redox electrodialysis process for pH-driven separations, and untangle the concentration- and current-dependent membrane transport phenomena that limit the process efficiency.Engineering and Applied Sciences - Engineering Science
Stigma and the Social Safety Net
Get PDFStigma features prominently in debates about the social safety net, but empirically disentangling its role has left open many questions about whether it is a meaningful—or movable—barrier to take-up. Through four nationally representative studies (N = 11,164) and a new four-dimensional validated scale, we quantify the role that stigma plays in shaping take-up (1) directly, by impacting beneficiary behavior, and (2) indirectly, by influencing program design. We find that a one standard deviation (SD) increase in stigma is associated with a 9-19 percentage point decrease in willingness to apply for benefits among low-income respondents. It also predicts a 0.08-0.40 SD increase in society’s preferences for policies and program design features that could reduce program access. In both cases, we show that stigma explains more of the variation in policy preferences than any individual respondent characteristic, including political ideology. Notably, program design causally impacts stigma in competing ways: more expansive eligibility criteria reduce stigma, while implementation designs that would simplify access increase stigma. Together, these findings suggest that stigma should be considered both an individual and structural barrier to participation in the social safety net, where it both shapes and is shaped by policy design choices.Version of Recor